Stage IV neuroblastoma (NB) in children >18 months of age is an aggressive disease that often recurs after successful induction therapy. Despite intensive treatment regimens, survival rates are unacceptable. Ganglioside GD2 expressed on NB is an ideal target for antibody-dependent cellular cytotoxicity. Anti-GD2 monoclonal antibody (mAb) immunotherapy in stage IV patients with minimal residual disease significantly prolongs progression free survival (PFS) and increased overall survival (OS); however, 37% of these stage IV patients relapse within 2 years. Unfortunately, clinical efficacy of anti-GD2 mAbs has been limited by toxicities. There is a need for novel and innovative approaches that increase anti-tumor activity without increasing toxicities. We recently showed that anti-CD3 activated T cells (ATC) armed with murine GD2Bi (anti-CD3 x anti-GD2 bispecific antibody) and ATC armed with anti-CD3 x anti-GD2 (humanized version of 3F8) exhibit enhanced in vitro and in vivo potency against NB cells. Delayed tumor growth and improved survival was seen in a NB xenograft model following infusions of hu3F8Bi armed ATC. In a phase I clinical trial for metastatic breast, multiple infusions of anti-CD3 x anti-Her2 bispecific antibody (Her2Bi) armed ATC in combination with low dose IL-2 and granulocyte-macrophage colony stimulating factor (GM-CSF) in women with heavily pretreated stage IV metastatic 0-3+ Her2 positive breast cancer were safe, induced anti-breast cancer cytotoxicity, anti-breast cancer antibodies, serum Th1 cytokine patterns, and IL-12 levels. Surprisingly, median OS was 36 months with 1 very good partial response in a patient with liver metastases and 12 patients with stable disease out of 22 evaluable pts at 15 weeks after starting immunotherapy. We hypothesized that hu3F8Bi armed ATC will enhance anti-NB cytotoxicity and vaccinate the pts against their own NB antigens. To test this hypothesis, we propose the following specific aims: 1) Submit a IND, produce GD2Bi (hu3F8Bi), and perform a phase I dose-escalation protocol in patients with recurrent/refractory NB and GD2+ tumors to determine the MTD for hu3F8Bi armed ATC infused twice a week for 8 infusions in combination with daily low dose IL-2 (300,000 IU/m2/day) in a standard 3 + 3 dose escalation schema with 40, 80, and 160 x 106 cells/kg/infusion dose levels; 2) Conduct a phase II clinical trial in 22 patients to explore efficacy and confirm the toxicity profile of hu3F8Bi armed ATC using the MTD; 3) Sequentially monitor immune responses directed at NB (cytotoxicity, IFN? EliSpots), phenotyping, cytokine patterns, and antibodies to NB; 4) Assess survival and persistence of armed ATC in the blood and tumor biopsies to confirm trafficking of armed ATC to tumor; and 5) Conduct an exploratory study of [18F] FDG PET/CT in 10 selected patients with PET/CT measurable soft tissue and skeletal lesions to evaluate trafficking of armed ATC. If successful, these studies will provide a shift in the treatment paradigm for NB where cytotoxicity would not only reduce minimal residual disease but also vaccinate the patients against their own tumor resulting in significant improvement in PFS and OS.